• Không có kết quả nào được tìm thấy

Thư viện số Văn Lang: Secondary Analysis of Electronic Health Records

N/A
N/A
Nguyễn Gia Hào

Academic year: 2023

Chia sẻ "Thư viện số Văn Lang: Secondary Analysis of Electronic Health Records"

Copied!
77
0
0

Loading.... (view fulltext now)

Văn bản

But "big data" in healthcare has become coated in "Silicon Valley Disruptionese," the language in which Silicon Valley spins hype into startup gold and fills it with grandiose promises to lure investors and early adopters. The extremely high granularity of the data allows for complicated analysis of complex clinical problems.

Fig. 2.1 Basic overview of the MIMIC database
Fig. 2.1 Basic overview of the MIMIC database

Access and Interface

PCORnet

Included Variables

Access and Interface

Open NHS

Included Variables

Access and Interface

The National Institute for Health Research (NIHR) Clinical Research Network (CRN) has produced and implemented an online tool known as the Open Data Platform. Clinical Research Network staff can access the Open Data platform and determine the number of patients recruited for research studies in a given hospital, as well as the research conducted at that hospital.

Other Ongoing Research

VistA

The large collection of information contained in the database will enable a wide range of researchers to improve clinical care in many domains. Strengths of the data include the ability to track patients across the United States, as well as from inpatient to outpatient settings.

NSQUIP

Scott DJ, Lee J, Silva I et al (2013) Access to the public MIMIC-II intensive care relational database for clinical research. Clinicians, researchers, and data scientists will need to navigate numerous challenges facing big data analytics—including system interoperability, data sharing, and data security—to realize the full potential of EHR and big data-based studies.

Introduction

Electronic health records (EHR) are increasingly useful for conducting secondary observational studies with power that rivals randomized controlled trials. Secondary analysis of EHR data can inform large-scale health system choices (eg, pharmacovigilance) or point-of-care clinical decisions (eg, medication selection).

Challenges in Secondary Analysis of Electronic Health Records Data

Such reports highlight the need for prudence amid the exuberance when using aggregated electronic health data for big data analytics—such use comes with an ethical responsibility to protect population-level and personal data from activity criminal and other nefarious purposes. No figures exist for the creation of a large healthcare data center, and these figures would be variable anyway, depending on the scale and type of data.

Opportunities in Secondary Analysis of Electronic Health Records Data

These collaborative enterprises will have transparent governance structures and standards for data access, allowing study validation and ongoing peer review of published and unpublished work [15], and mitigating the effects of selection bias and confounding in any single study [17]. New standards for data sharing may need to be implemented for institutions to be truly comfortable with sharing records, but within institutions and existing research collaborations, safe practices for data security can be implemented, and greater collaboration encouraged through standardization of data entry and storage.

Secondary EHR Analyses as Alternatives to Randomized Controlled Clinical Trials

There was a substantial difference in the magnitude of effect sizes in one third of the comparisons, reaching statistical significance in one case [27]. Finally, in the case of databased observational studies, it becomes much more feasible to improve and replicate, or simply repeat, studies if this is deemed necessary to investigate accuracy, heterogeneity of effects, and new clinical insights.

Demonstrating the Power of Secondary EHR Analysis

The results of Ioannidis et al. [24] confirms a more recent Cochrane meta-analysis, which found no significant differences in effect estimates between RCTs and observational studies, regardless of observational study design or heterogeneity. For example, the highly publicized results of the Kaiser study on Vioxx substantiated earlier suspicions of an association between celecoxib and an increased risk of serious coronary heart disease [30].

A New Paradigm for Supporting Evidence-Based Practice and Ethical Considerations

URL: http://www.fiercehealthit.com/story/big-datas-biggest- healthcare-challenge-making-sense-it-all/2015-05-04. Kitsios GD, Dahabreh IJ, Callahan S et al (2015) Can we trust observational studies using propensity scores in the critical care literature.

Use Case Examples Based on Unavoidable Medical Heterogeneity

Clinical Work fl ow, Documentation, and Decisions

A clinical issue such as infection or vascular occlusion affects the patient's condition. Transfer control involves transmitting disturbances directly to the sensor without first affecting the patient's condition.

Levels of Precision and Personalization

Based on these analyzes which are continuously performed by the system and updated as the user starts to enter a note, a series of problems are identified and presented to the user through EMR display. While the user interface of the monitor is currently only one for adjusting the monitored channels and the alarm settings, the user interface will also be increasingly rich so that the user, for example with the right credentials, can access, edit and annotate the EHR from a bedside or central monitor, or add information directly to the monitor to calibrate the monitoring process.

Coordination, Communication, and Guidance Through the Clinical Labyrinth

The range of those affected by the challenges of the current healthcare system is wide. As we have seen at the level of individual patients, too much data is not useful if it is not processed, analyzed, placed in the right context and available to the right people in the right place at the right time.

Safety and Quality in an ICS

Optimal results are desirable, but not necessarily so, in the broadest sense of the term. Predictive analytics can be developed as system elements to prospectively inform users of security and quality threats [19-21].

Fig. 4.4 Control loop depicting a data-driven safety system. A clinical safety issue affects the state of the patient
Fig. 4.4 Control loop depicting a data-driven safety system. A clinical safety issue affects the state of the patient

Conclusion

Celi LA, Mark RG, Stone DJ, Montgomery R (2013) “Big data” in the ICU: closing the data loop. Celi LA, Marshall JD, Lai Y, Stone DJ, Physician documentation and decision making in the digital age.

Fig. 4.5 Information Architecture of an Ideal Care System. This diagram integrates the concepts described in this chapter depicting data driven care systems, safety systems, along with connection and coordination of patient data across multiple modalities
Fig. 4.5 Information Architecture of an Ideal Care System. This diagram integrates the concepts described in this chapter depicting data driven care systems, safety systems, along with connection and coordination of patient data across multiple modalities

The Vision

MIMIC is a medical information market for intensive care and consists of several comprehensive data streams in the intensive care environment, in high levels of richness and detail, supporting complex signal processing and clinical inquiry that could allow early detection of complex problems, provide useful guidance on therapeutic interventions and in ultimately lead to improved patient outcomes. Such systems would allow early detection of complex problems, provide useful guidance on therapeutic interventions, and ultimately lead to improved patient outcomes.

Data Acquisition

Clinical Data

Physiological Data

Death Data

Data Merger and Organization

Data Sharing

Updating

Support

Lessons Learned

Future Directions

Non-clinical factors make a significant contribution to an individual's health, and providing these data to clinicians can inform context, counseling and treatments.

Introduction

Non-clinical Factors and Determinants of Health

Behavioral factors such as physical activity, diet, smoking and alcohol consumption are strongly related to the epidemic of obesity [2]. Also included are social stressors such as crime, violence and physical ailments, as well as others [4].

Increasing Data Availability

The fact that much non-clinical data, especially urban data, is geographically localized allows physicians to view patients' health in a broader perspective. Connections and solutions become more visible by linking non-clinical data to EHR at the public health and urban planning level.

Integration, Application and Calibration

There are further limitations due to data resolutions, which refer to the level of detail of data in space, time or theme, especially the spatial dimension of the data [16]. This method reports the first three digits of the patient's zip code, while omitting the last two digits [18].

Table 6.2 A tidy dataset that contains a readily machine-readable format of the data in Table 6.1 Patient ID Place Date (MM/DD/YYYY) Pressure (mmHg) Cycle
Table 6.2 A tidy dataset that contains a readily machine-readable format of the data in Table 6.1 Patient ID Place Date (MM/DD/YYYY) Pressure (mmHg) Cycle

A Well-Connected Empowerment

Challenges remain in protecting confidentiality at a single patient level and determining the applicability of macroscopic data to the single patient. As geographic de-identified health information becomes publicly available, an electronic medical record will be able to download this information from the cloud, apply it to the patient's zip code, gender, age and sexual preferences (if documented), and alert/guide the clinic that would decide whether an intervention is required based on a calculated risk of getting an STD.

Conclusion

Kheirbek I, Wheeler K, Walters S, Kass D, Matte T (2013) PM2.5 and ozone health impacts and disparities in New York City: sensitivity to spatial and temporal resolution. Jeffery N, McKelvey W, Matte T (2015) Using tracking infrastructure to support public health programs, policies, and emergency response in New York City.

Introduction

While EHR data is extensive and the analyzes are powerful, it is essential to fully understand the biases and limitations introduced when used in healthcare research. Fundamentally, healthcare research places the healthcare system under the microscope as the organism of inquiry.

The Rise of EHRs in Health Services Research .1 The EHR in Outcomes and Observational Studies

The EHR as Tool to Facilitate Patient Enrollment in Prospective Trials

Despite the power of the EHR to retrospectively conduct health services and outcomes research, the gold standard in research remains prospective and randomized trials. 21] suggested using the EHR to simultaneously generate clinical trial alerts, especially in commercial EHRs such as Epic to leverage the EHR in a point-of-care strategy.

The EHR as Tool to Study and Improve Patient Outcomes

This strategy could accelerate enrollment, although it must be weighed against the risk of losing patient confidentiality, an ongoing tension between patient care and clinical trial enrollment [ 22 ].

How to Avoid Common Pitfalls When Using EHR to Do Health Services Research

Step 1: Recognize the Fallibility of the EHR

Table 7.1 highlights examples of different diagnoses and possible sources of data that may or may not be present in all patients. Proposed Research Study: Researchers will need to sort out which patients were exposed to antipsychotics and which were not.

Step 2: Understand Confounding, Bias, and Missing Data When Using the EHR for Research

As a result, antipsychotics would appear to be associated with a higher risk of in-hospital mortality and the use of hospital resources, not because of the independent effect of the drug, but rather because of confounding by indication. Missing data or unevenly sampled data collected as part of the EHR creates its own complex set of challenges for health services research.

Future Directions for the EHR and Health Services Research

Ensuring Adequate Patient Privacy Protection

Multidimensional Collaborations

Fundamental to the feasibility of multidimensional collaborations is the ability to ensure data accuracy at scale and integrate it across multiple health data technologies and platforms.

Conclusion

Post AR, Kurc T, Cholleti S, Gao J, Lin X et al (2013) Analytical information warehouse. AIW): a platform for analytics using electronic health records. Byrne CM, Mercincavage LM, Pan EC, Vincent AG, Johnston DS et al (2010) Value from.

Introduction

There are numerous examples [3,4] in clinical medicine where observational data have been used to inform clinical decision-making, only to be ultimately refuted and potentially cause harm in the meantime. The unrecognized effect of an additional variable related to the primary exposure that influences the outcome of interest is known as confounding.

Confounding Variables in Big Data

The Obesity Paradox

As obesity is typically defined by body mass index (BMI) at ICU admission, it is possible that unrecognized influences on preadmission body weight, which independently influence outcome, may be the true cause of this paradoxical association. Contrast that with another such patient who developed cachexia from metastatic cancer and lost 30 pounds before presenting to the ER.

Selection Bias

Thus, this fluid accumulation would prompt the emergency room team to admit the patient to the ICU rather than the general medicine ward. That patient's BMI would have dropped significantly in the weeks before the illness, and his poor prognosis and illness could have led to an ICU admission, where his prognosis would have been poor.

Uncertain Pathophysiology

There has been an abundance of studies in the critical care literature linking renal function to a host of outcomes [12,13]. Regarding the association of renal function with cardiovascular mortality, there are many determinants of cardiac function that simultaneously and independently affect both serum creatinine.

Fig. 8.2 Concept map of the association of renal function and cardiovascular mortality revealing more of the confounding in fl uences
Fig. 8.2 Concept map of the association of renal function and cardiovascular mortality revealing more of the confounding in fl uences

Conclusion

Apel M, Maia VPL, Zeidan M, Schinkoethe C, Wolf G, Reinhart K et al (2013) End-stage renal disease and outcome in a surgical intensive care unit. On behalf of the working group on sepsis-related problems in the European Society of Intensive Care Medicine.

Hình ảnh

Fig. 2.1 Basic overview of the MIMIC database
Table 4.1 Clinical use cases with pertinent clinical and data objectives Clinical use case Clinical objective(s) Data objectives Outpatient in state of
Fig. 4.4 Control loop depicting a data-driven safety system. A clinical safety issue affects the state of the patient
Fig. 4.5 Information Architecture of an Ideal Care System. This diagram integrates the concepts described in this chapter depicting data driven care systems, safety systems, along with connection and coordination of patient data across multiple modalities
+6

Tài liệu tham khảo

Tài liệu liên quan

Read the following passage and mark the letter A, B, C, or D on your answer sheet to indicate the correct word or phrase that best fits each of the numbered blanks.. Many students

In his research, he combines his expertise in machine learning and critical care to generate new clinical evidence and build the next generation of clinical tools such as decision